MRI无创定量测量腹腔脂肪体积预测2型糖尿病  

Visceral abdominal adiposity tissue volume quantification using noninvasive MRI in prediction of type 2 diabetes

作　　者：王猛[1] 罗宴吉[1] 蔡华崧[1] 高进云[2] 贾应梅 李子平[1] 孟悛非[1] 冯仕庭[1] 

机构地区：[1]中山大学附属第一医院放射诊断科,广东广州510080 [2]深圳市中医院放射影像科,广东深圳518033

出　　处：《中国医学影像技术》2017年第12期1844-1849,共6页Chinese Journal of Medical Imaging Technology

基　　金：国家自然科学基金(81571750);广东省自然科学基金(2015A030313043;2014A030311018);广东省科技计划项目(2014A020212125)

摘　　要：目的探讨MRI无创定量测量腹腔脂肪体积(VAT)在预测2型糖尿病中的作用。方法收集2型糖尿病(T2DM)患者15例、糖耐量异常(IGT)患者17例、正常人群(糖耐量正常,NGT)16名。所有试验者均接受上腹部MRI I-DEAL-IQ序列检查,利用后处理工作站在T1WI IDEAL-IQ序列的脂肪分量图上分别测量并计算L2、L3范围所对应的腹腔脂肪体积(VATV L2、VATV L3),并计算两者总量(total VATV);利用ROC曲线及Logistic回归方程评估各部位脂肪含量预测T2DM的可能性。结果 T2DM组VATV L2、VATV L3、total VATV均高于IGT组与NGT组(P均<0.05);IGT与NGT组VATV L2、VATV L3、total VATV差异无统计学意义(P>0.05)。VATV L2>460.34 ml时,诊断T2DM的敏感度为73.33%,特异度为75.76%,准确率为75.00%;VATV L3>429.46 ml时诊断T2DM的敏感度为86.67%,特异度为72.73%,准确率为77.08%;total VATV>887.83ml时诊断T2DM的敏感度为86.67%,特异度为72.73%,准确率为77.08%。二分类Logistic回归方程分析显示,可纳入的变量为VATV L3(P=0.01,OR=1.01),判断T2DM的敏感度为80.00%,特异度为88.20%,总体准确率为84.40%。结论 MRI定量腹腔脂肪体积提供一种有效的无创性生物学指标预测T2DM的发生,VATV L3是较好的预测因子。Objective To investigate the feasibility of utilizing visceral abdominal adiposity tissue （VAT） volume quantification using MRI to predict type 2 diabetes mellitus （T2DM）. Methods Forty-eight subjects including 15 T2DM （T2DM group）, 17 impaired glucose tolerance （IGT, IGT group） and 16 normal glucose tolerance （NGT, NGT group） were enrolled in this study. All subjects underwent upper abdominal iterative decomposition of water and fat with echo asymmetry and least square estimation-image quantification （IDEAL-IQ） MRI scanning. VAT volume of the second and third lumber vertebral body ranges （VATV L2, VATV L3）, sum of VATV L2 and L3 （total VATV）, hepatic and pancreatic fat were measured in fat fraction mapping of T1WI IDAEL-IQ sequence on post-processing workstation. The accuracy of predicting T2DM using VAT was evaluated by Logistic regression equation via ROC curve. Results The mean of VATV L2, VATV L3 and total VATV in T2DM group were significantly higher than those of IGT group and NGT group （P〈0.05）, while there were no significant difference of these metrics between IGT group and NGT group （P〉0.05）. Taking 460.34 ml as the cut-off value for VATV L2 to predict T2DM, sensitivity was 73.33%, specificity was 75.76% and accuracy was 75.00%, respectively. Taking 429.46 ml as the cut-off value for VATV L3 to predict T2DM, sensitivity was 86.67%, specificity was 72.73% and accuracy was 77.08%, respectively. Taking 887.83 ml as the cut-off value for total VATV to predict T2DM, the sensitivity, specificity and accuracy were 86.67%, 72.73% and 77.08%, respectively. Only VATV L3 was enrolled by Logistic regression equation （P=0.01, OR=1.01）, and the sensitivity, specificity and total accuracy of prediction for T2DM were 80.00%, 88.20%, and 84.40%, respectively. Conclusion It is feasible to utilize VAT volume quantification with MRI to predict T2DM. VATV L3 is a better predictor.

关 键 词：磁共振成像 糖尿病 2型 糖耐量试验 

分 类 号：R814.42[医药卫生—影像医学与核医学] R816.5[医药卫生—放射医学]